smra error

[adaptivegenome]

I sent you an email. Thank you so much for replying! I am very excited to try the C version.

[nilshomer]

I'm having a similar problem. I tried 4G, my average coverage is 30X and it dies whether I provide a region list or not. Any ideas? Did you guys figure out the problem above?

Java, java, java... that is why I am re-writing it in C. PM me if you are interested in a C-version, otherwise I would be happy to help at [email protected] or within a PM.

[adaptivegenome]

I'm having a similar problem. I tried 4G, my average coverage is 30X and it dies whether I provide a region list or not. Any ideas? Did you guys figure out the problem above?

[nilshomer]

Hi
I tried using 10G without using any RANGE file (as it gives me different kind of problem), after a while smra crashes...

The ranges are still not sorted. You could write a quick perl script to sort them by coordinate. What is the stacktrace when it crashes? I would be very interested in trying to help you with this problem. Feel free to PM or email me (or the help list).

[zee]

I think you should mail the author your version number, STDERR report, etc.

What would happen if you tried it with
1) only chr1 ranges ?
2) only chr10 ranges ?

[seq_GA]

You don't need to adjust the heap size of SRMA, though add more memory with java ("java -Xmx2G -jar srma-0.1.6.jar"). The ranges must be in sorted order. Check the line at which it complains.

Hi
I tried using 10G without using any RANGE file (as it gives me different kind of problem), after a while smra crashes...

[seq_GA]

Hi Zee,
I tried that also.
Its always breaking down when there is a trend in change of nimbers as below: When the nums change from 9 digit number to 5 digit number..

chr1    247118921       247119161
chr1    247174632       247174752
[B]chr1    247177404       247179204
chr10   82880   83000[/B]
chr10   83041   84088
chr10   84549   84669

[zee]

I think you should change your sort to

sort -k 1,1 -k 2n,2 range.txt -o range.txt

[seq_GA]

Hi,
Thx for the suggestion. I tried using java -Xmx2G -jar and it crashed again. Now I am trying with 10G. Let me see whether it gets completed

Regarding my range file, its line 272 is as in the bold line. And it is tab delimited data.

chr10 102296881 102297001
chr10 102297801 102297921
chr10 1024383 1024503
chr10 102495968 102496088
chr10 102499456 102499696

I sorted my region of interest as

sort range.txt > range_sorted.txt

If I remove that line and move ahead it it works. This looks like sorting problem. Any suggestions. Thanks. If I fix this may be I will gain some running time. Thanks.

[nilshomer]

Hi,
The mean coverage is about 110. I am yet to try the heap size adn ranges file which is giving trouble.

You don't need to adjust the heap size of SRMA, though add more memory with java ("java -Xmx2G -jar srma-0.1.6.jar"). The ranges must be in sorted order. Check the line at which it complains.

[seq_GA]

Hi,
The mean coverage is about 110. I am yet to try the heap size adn ranges file which is giving trouble.

[nilshomer]

How much should I specify for MAX_HEAP_SIZE?

I usually set my java memory heap quite high for this task e.g.

That gives it about 4Gb of RAM to work with. Perhaps bump yours up to 4/8Gb.

This is an interesting question. I usually set it to ~1.25GB for 30x coverage, but it could require more if there is significantly more coverage. Let me know your mean coverage and heap size that works so I can enlighten myself.

[zee]

I usually set my java memory heap quite high for this task e.g.

#!/bin/sh
# filename = srma & is executable

java -Xmx4000m -jar /export/opt/srma/build/jar/srma-0.1.6.jar $*

That gives it about 4Gb of RAM to work with. Perhaps bump yours up to 4/8Gb.

[seq_GA]

Hi,

I am getting the following error while using smra.

 java -jar /srma/srma-0.1.6/srma-0.1.6.jar I=KLM_sorted.bam O=KLM_sorted_smra.bam R=/Genome/hg18/hg18.fa VALIDATION_STRINGEN
CY=SILENT
[Thu Jun 24 16:12:12 SGT 2010] srma.SRMA INPUT=KLM_sorted.bam OUTPUT=KLM_sorted_smra.bam REFERENCE=/Genome/hg18/hg18.fa VALIDATION_STRINGENCY=SILENT    OFFSET=20 MIN_MAPQ=0 MINIMUM_ALLELE_PROBABILITY=0.1 MINIMUM_ALLELE_COVERAGE=3 MAXIMUM_TOTAL_COVERAGE=100 CORRECT_BASES=false USE_SEQUENCE_QUALITIES=true QUIET_STDERR=false MAX_HEAP_SIZE=8192 MAX_QUEUE_SIZE=65536 NUM_THREADS=1 TMP_DIR=/tmp/userrig VERBOSITY=INFO QUIET=false COMPRESSION_LEVEL=5 MAX_RECORDS_IN_RAM=500000
Allele coverage cutoffs:
coverage:  1    minimum allele coverage: 0
coverage:  2    minimum allele coverage: 0
coverage:  3    minimum allele coverage: 0
coverage:  4    minimum allele coverage: 1
coverage:  5    minimum allele coverage: 1
coverage:  6    minimum allele coverage: 1
coverage:  7    minimum allele coverage: 2
coverage:  8    minimum allele coverage: 2
coverage:  9    minimum allele coverage: 3
coverage: >9    minimum allele coverage: 3
Records processsed: 65534 (last chr10:1904158-1904236)                                                      Exception in thread "Thread-2" java.lang.OutOfMemoryError: Java heap space
[Thu Jun 24 16:16:07 SGT 2010] srma.SRMA done.
Runtime.totalMemory()=954466304
Exception in thread "main" java.lang.OutOfMemoryError: Java heap space
        at java.util.LinkedList.addBefore(LinkedList.java:795)
        at java.util.LinkedList.add(LinkedList.java:215)
        at srma.ThreadPoolLinkedList.add(ThreadPoolLinkedList.java:19)
        at srma.SRMA.processToAddToGraphList(SRMA.java:485)
        at srma.SRMA.doWork(SRMA.java:272)
        at net.sf.picard.cmdline.CommandLineProgram.instanceMain(CommandLineProgram.java:150)
        at srma.SRMA.main(SRMA.java:94)

How much should I specify for MAX_HEAP_SIZE?